CMP364 Machine Learning
Machine Learning notes
9 chapter notes, in syllabus order. Each starts with the key points.
Unit 1
ML Basics: Definitions, Types, and Real-World ApplicationsUnit 1 of Machine Learning introduces core concepts like supervised vs. unsupervised learning, key ML tasks (classification, regression, clustering), and real-world applications in apps like eSewa, Pathao, and NEPSE. It covers ML vs. AI, problem formulation, and ethical considerations.14 min readUnit 2
Data Preprocessing: Cleaning, Transforming & Feature EngineeringUnit 2 of Machine Learning covers essential techniques to prepare raw data for modeling—handling missing values, scaling, encoding, feature selection, and dimensionality reduction—with real-world applications in Nepalese tech (e.g., Khalti fraud detection, Pathao route optimization) and global platforms (Google’s recom12 min readUnit 3
Linear & Logistic Regression: Models, Math & ApplicationsUnit 3 of Machine Learning covers linear regression (predicting continuous outputs) and logistic regression (binary classification), including their mathematical foundations, assumptions, real-world applications, and implementation steps with Python-style pseudocode.12 min readUnit 4
Decision Trees & Ensembles: Splits, Pruning, Boosting & BaggingUnit 4 of Machine Learning covers how decision trees classify data by recursive splitting, how to prune them to avoid overfitting, and how ensembles like Random Forest and AdaBoost combine weak learners into strong models—with real-world examples from eSewa fraud detection and Pathao route optimization.9 min readUnit 5
Support Vector Machines: Kernels, Optimization & ClassificationUnit 5 of Machine Learning explores Support Vector Machines (SVMs), covering their mathematical foundations, kernel tricks, hyperplane optimization, and real-world applications in classification and regression. Learn how SVMs maximize margin, handle non-linear data, and compare with other models.16 min readUnit 6
Bayesian Learning: Probability, Naive Bayes, MAP, EM, and ApplicationsUnit 6 of Machine Learning explores Bayesian inference, conditional probability, Naive Bayes classifiers, Maximum A Posteriori (MAP) estimation, and the Expectation-Maximization (EM) algorithm, with real-world applications in spam filtering, medical diagnosis, and recommendation systems.8 min readUnit 7
Clustering: Algorithms, Applications & EvaluationUnit 7 of Machine Learning explores unsupervised learning techniques for grouping similar data points, covering centroid-based (K-means), density-based (DBSCAN), hierarchical, and spectral clustering, with real-world applications in recommendation systems, anomaly detection, and image segmentation.12 min readUnit 8
Dimensionality Reduction: PCA, t-SNE, Autoencoders & ApplicationsUnit 8 of Machine Learning explores techniques to reduce feature space dimensions while preserving data structure, covering PCA, t-SNE, autoencoders, and their real-world applications in data compression, visualization, and noise reduction.19 min readUnit 9
Neural Networks & Deep Learning: Architectures, Training, and ApplicationsUnit 9 of Machine Learning explores neural networks (MLPs, CNNs, RNNs), deep learning architectures, backpropagation, activation functions, and real-world applications like image recognition and NLP, with worked examples and exam-focused insights.12 min readUnit 10
Model Evaluation · note coming